Image enhancement method based on bionic adaptive memristive neural network

A neural network and image enhancement technology, applied in the field of image processing, can solve problems such as the lack of bionic considerations, the failure to simulate the global and local adaptive adjustment characteristics of the human visual system, and the difficulty of obtaining ideal results with fixed templates, so as to facilitate hardware acceleration, It is beneficial to large-scale circuit integration and enhances the effect of image enhancement advantages

Active Publication Date: 2022-03-01
SOUTHWEST UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Cellular neural network has a simple local interconnection structure and high-speed parallel processing capability. It is the basic model for constructing artificial retina and can be applied to image enhancement in image processing in machine vision. However, the existing image enhancement methods are still There are some deficiencies, for example, when dealing with actual complex images, it is difficult to achieve ideal results by using fixed templates; moreover, it fails to simulate the global and local adaptive adjustment characteristics of the human visual system, and lacks bionic considerations.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image enhancement method based on bionic adaptive memristive neural network
  • Image enhancement method based on bionic adaptive memristive neural network
  • Image enhancement method based on bionic adaptive memristive neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, and are not intended to limit the invention.

[0047] Cellular neural network is a large-scale parallel analog information processing array. The unit cells in the network have the same structure and regular local connections between cells. It is one of the most suitable neural network models for large-scale integrated circuits. Cellular neural networks are widely used in many fields such as image processing, object detection, and pattern recognition, and are also the model basis for constructing artificial retinas. In the cellular neural network, the template is the key factor to determine the output of the network, and di...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an image enhancement method based on a bionic adaptive memristive cellular neural network, which integrates an adaptive three-Gaussian model simulating the human visual mechanism, a cellular neural network and a nano-memristor, in the typical adaptive three-Gaussian model Based on this, a linear adaptive three-Gaussian model that is easy to implement in hardware is proposed. Using the Gaussian kernel function and the image processing characteristics of the cellular neural network, it overcomes the limitations of fixed image enhancement templates and pure mathematical algorithms, and greatly enhances the advantages of image enhancement. Its enhancement effect is closer to that of the human eye; at the same time, the hardware design based on memristor is conducive to large-scale circuit integration, and it is convenient to realize hardware acceleration and chip-based end-side real-time image processing.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image enhancement method based on a bionic adaptive memristive neural network. Background technique [0002] Image enhancement can emphasize the interesting features on the image and suppress the unwanted features, so as to enhance the useful information of the image and improve the image quality. Cellular neural network has a simple local interconnection structure and high-speed parallel processing capability. It is the basic model for constructing artificial retina and can be applied to image enhancement in image processing in machine vision. However, the existing image enhancement methods are still There are some deficiencies, for example, when dealing with actual complex images, it is difficult to achieve ideal results with fixed templates; moreover, it fails to simulate the global and local adaptive adjustment characteristics of the human visual system, and lacks...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06T5/00
CPCG06T5/007G06T5/001G06T2207/10004G06T2207/10024G06T2207/20081G06T2207/20084
Inventor 胡小方郑雅文周跃段书凯
Owner SOUTHWEST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap